CRACI: A Cloud-Native Reference Architecture for the Industrial Compute Continuum
Hai Dinh-Tuan · 2025
Read CRACI: A Cloud-Native Reference Architecture for the Industrial Compute Continuum by Hai Dinh-Tuan (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2509-07498.txt.
Introduction
Factories still run on rigid, layered control charts drawn decades ago, and those charts block modern cloud tools. Hai Dinh-Tuan proposes an alternative, CRACI, in a 2025 research paper written alone. CRACI is a cloud-native reference architecture for the industrial compute continuum, the mix of factory devices, edge servers, and clouds. It replaces fixed hierarchies with a decoupled, event-driven design.
Data can then flow flexibly between levels instead of climbing a strict ladder. Trust, governance, observability, and lifecycle management are built in as founding pillars, not add-ons. Dinh-Tuan validates the design two ways. He compares it against established standards and models, and he reuses performance data from published smart manufacturing projects. Both checks suggest CRACI can support scalable, interoperable industrial systems better than the legacy models.
Catalog record
| CREATOR | Hai Dinh-Tuan |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 36 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2509-07498
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
CRACI: A Cloud-Native Reference Architecture for the Industrial Compute Continuum Hai Dinh-Tuan Abstract The convergence of Information Technology (IT) and Operational Technology (OT) in Industry 4.0 exposes the limitations of traditional, hierarchical architectures like ISA-95 and RAMI 4.0. Their inherent rigidity, data silos, and lack of support for cloud-native technologies impair the development of scalable and interoperable industrial systems. This paper addresses this issue by introducing CRACI, a Cloud-native Reference Architecture for the Industrial Compute Continuum. Among other features, CRACI promotes a decoupled and event-driven model to enable flexible, non-hierarchical data flows across the continuum. It embeds cross-cutting concerns as foundational pillars: Trust, Governance & Policy, Observability, and Lifecycle Management, ensuring quality attributes are core to the design. The proposed architecture is validated through a two-fold approach: (1) a comparative theoretical analysis against established standards, operational models, and academic proposals; and (2) a quantitative evaluation based on performance data from previously published real-world smart manufacturing implementations. The results demonstrate that CRACI provides a viable, state-of-the-art architecture that utilizes the compute continuum to overcome the structural limitations of legacy models and enable scalable, modern industrial systems. I Introduction The Fourth Industrial Revolution (Industry 4.0) is driving an unprecedented convergence of physical production systems and digital intelligence. Modern industrial environments demand architectures that support scalable, low-latency analytics, autonomous control, and seamless integration across the entire Compute Continuum (CC), from edge devices on the factory floor to centralized cloud platforms. However, the models that current industrial systems rely on have proven inadequate for fully harnessing the benefits of this paradigm. Established standards such as ISA-95 and the Reference Architecture Model for Industry 4.0 (RAMI 4.0) , while valuable for structuring enterprise-control systems, are derived from a rigid, hierarchical view. Their top-down structure creates data silos, limits real-time responsiveness, and fails to natively support the decentralized, event-driven, and microservice-based patterns that define modern cloud-native systems. This fundamental mismatch creates a significant architectural gap, resulting in inflexible, monolithic systems that are difficult to maintain, operate, and scale. While more recent industry-driven operational models like FIWARE and NOA partly address aspects of this problem, they often remain either too abstract, too technologically prescriptive, or incomplete, lacking a unified operational model for managing distributed services across heterogeneous environments. Similarly, academic proposals often present conceptual models that, while insightful, often lack the practical implementation details and operational guidance necessary for real-world adoption. To address this critical gap, this work introduces CRACI: a Cloud-native Reference Architecture for the Compute Continuum in Industry 4.0. Instead of a rigid pyramid, CRACI proposes a decoupled, service-oriented system built on two core patterns: a hub-and-spoke model centered on a Digital Twin repository and an event-driven communication backbone. This design overcomes data silos and enables flexible, non-hierarchical interactions between components. A key novelty of the proposed architecture is the formal integration of four Foundational Pillars : Trust, Governance & Policy, Observability, and Lifecycle Management, which embed critical quality attributes directly into the system design. This approach ensures the architecture is not only aligned with cloud-native best practices but also maintains a balance between high-level abstraction (for general applicability) and detailed guidance (for concrete implementation). The primary contributions of this work are therefore threefold: • An analytical review of Industry 4.0 architectural standards and models, highlighting critical gaps in operational and governance concerns. • The design process of CRACI, resulting in a novel cloud-native reference architecture that resolves the identified gaps through a data-centric design and the formalization of four Foundational Pillars. • A qualitative and quantitative evaluation demonstrating how CRACI’s design systematically addresses the limitations of existing models like ISA-95 or RAMI 4.0. The remainder of this work is structured as follows: Section II reviews related work in industrial reference architectures. Section III presents the design methodology while Section IV introduces the CRACI architecture in detail. Section V provides the theoretical comparative analysis, followed by the quantitative evaluation in Section VI. Finally, Section VII concludes the work by summarizing the contributions and outlining directions for future work. II Related Work To position our contribution, we review three classes of existing work, which are formal standards, industry-driven frameworks, and academic architectures to identify the specific gaps that motivate the design of CRACI. II-A Formal Industry Standards The ISA-95 model [ 1 ] has long served as the foundational model for integrating enterprise and control systems. Based on the Purdue Enterprise Reference Architecture , it defines a five-level automation pyramid that organizes functions from the physical process level (Level 0) up to business logistics (Level 4). While ISA-95 provides essential structural clarity, its rigid, top-down hierarchy is fundamentally ill-suited for modern, event-driven environments. It was designed before the era of cloud computing, leading to the creation of data silos, which severely impede the kind of flexible, cross-layer data access required for advanced analytics and distributed intelligence in Industry 4.0. As a direct response to the needs of Industry 4.0, Germany’s Plattform Industrie 4.0 introduced the Reference Architecture Model for Industry 4.0 (RAMI 4.0) [ 2 ] . RAMI 4.0 extends ISA-95 into a three-dimensional map as illustrated in Fig. 1 , with six Hierarchy Levels (based on IEC 62264 [ 3 ] and IEC 61512 [ 4 ] ), incorporating a Life Cycle & Value Stream axis (based on IEC 62890 [ 5 ] ) and an Architecture Layers axis to classify components. While RAMI 4.0 provides a valuable conceptual map for structuring assets and functions, it remains a highly abstract rather than operational. It offers little guidance on how to deploy, manage, or orchestrate distributed services, and it retains the hierarchical assumptions of ISA-95, which can be incompatible with direct cross-level interactions along the value chain found in modern manufacturing practices. Furthermore, critical cross-cutting concerns like governance and trust are not explicitly embedded within the model. Fig. 1: Reference Architecture Model Industrie 4.0 (RAMI 4.0) [ 2 ] . II-B Industry-Driven Operational Frameworks Several operational frameworks address these abstract limitations. For instance, the NAMUR Open Architecture (NOA) [ 6 , 7 ] offers an approach for brownfield environments by introducing a Monitoring and Optimization (M+O) domain that securely extracts data from the core process control domain as illustrated in Fig. 2 . However, NOA is intended as a transitional pattern, not a complete architecture. Moreover, due to its strong industrial origins, it treats the M+O domain merely as a conceptual block, lacking a concrete architectural model for the actual IT services and components that should operate within this domain. Fig. 2: NAMUR Open Architecture [ 6 , 7 ] . The FIWARE framework [ 8 ] offers a mature, data-centric platform centered around its NGSI-LD (Next Generation Service Interface-Linked Data) Context Broker. Despite its comprehensive design, its benefits are tied to a specific ecosystem of enablers , such as Orion Context Broker 1 1 1 https://github.com/telefonicaid/fiware-orion , Cygnus 2 2 2 https://github.com/telefonicaid/fiware-cygnus , Keyrock 3 3 3 https://keyrock-fiware.github.io/ , which creates a steep learning curve and risks technology lock-in. Furthermore, while critical concerns like trust and lifecycle management are acknowledged, they are handled externally instead of being embedded as core architectural principles. Lastly, the AIOTI Functional Model [ 9 ] , proposed by the EUCloudEdgeIoT.eu initiative, represents a significant effort to define key capabilities for